HTML
A complete, public company page at a stable canonical URL — the human-readable publication of the profile, built for people and browsers.
Schema.org, embedded as JSON-LD
The same verified facts, marked up as Schema.org structured data. It isn't a separate file to fetch — it's embedded directly inside the HTML page, which is how search engines and AI systems are meant to read it.
JSON
A machine-readable representation of the full profile — company details, products or services, sources and confidence — for software that reads the profile directly rather than parsing HTML.
llms.txt
A short, plain-text orientation file that points AI systems and agents to the profile's key facts and pages, so they don't have to guess where to look. It also links out to focused per-topic markdown files — pricing, products, services, addresses and more — so an agent that needs one specific fact can fetch exactly that file instead of the whole profile.
Sitemap
Published profiles are listed in a sitemap so search engines can discover and (re)crawl them as they're published or updated.